Using machine learning to detect system changes

A system comprising:

• a processor;
• and
• a memory comprising code stored thereon that, when executed, performs a method of detecting state changes in a system comprising:
• receiving a first neural network that is trained to detect when the system transitions into a first resulting state, wherein the system transitions into at least a first intermediate state prior to transitioning into the final resulting state;
• training the first neural network to recognize the first intermediate state using a first plurality of inputs denoting the system in the first intermediate state, wherein during said training the first neural network to recognize the first intermediate state, weights of the first neural network are adjusted at a plurality of different points in time after the first neural network has processed a portion of the first plurality of inputs;
• during said training the first neural network to recognize the first intermediate state, obtaining a plurality of sets of internal state information of the first neural network, each set of the plurality of sets denoting an internal state of the first neural network at one of the plurality of different points in time after the first neural network has processed a portion of the first plurality of inputs, wherein each set of the plurality of sets of internal state information includes adjusted weights captured at one of the plurality of different points in time during said training the first neural network;
• representing the plurality of sets of internal state information of the first neural network as a plurality of images;
• providing the plurality of images as inputs to a second neural network;
• and training the second neural network, using the inputs of the plurality of images representing the plurality of sets of internal state information, to detect the first intermediate state.

A non-transitory computer readable medium comprising code stored thereon that, when executed, performs a method of detecting state changes in a system comprising:

• receiving a first neural network that is trained to detect when the system transitions into a first resulting state, wherein the system transitions into at least a first intermediate state prior to transitioning into the final resulting state;
• training the first neural network to recognize the first intermediate state using a first plurality of inputs denoting the system in the first intermediate state, wherein during said training the first neural network to recognize the first intermediate state, weights of the first neural network are adjusted at a plurality of different points in time after the first neural network has processed a portion of the first plurality of inputs;
• during said training the first neural network to recognize the first intermediate state, obtaining a plurality of sets of internal state information of the first neural network, each set of the plurality of sets denoting an internal state of the first neural network at one of the plurality of different points in time after the first neural network has processed a portion of the first plurality of inputs, wherein each set of the plurality of sets of internal state information includes adjusted weights captured at one of the plurality of different points in time during said training the first neural network;
• representing the plurality of sets of internal state information of the first neural network as a plurality of images;
• providing the plurality of images as inputs to a second neural network;
• and
• training the second neural network, using the inputs of the plurality of images representing the plurality of sets of internal state information, to detect the first intermediate state.